Dataverse‎ > ‎

Congressional District Data


Scope of the data

The data set includes a wide range of economic, social, and geographic information for every U.S. congressional district, from 1943-1998.  The variables range from such basic information as size of each district (in square miles), the population, and the number unemployed (at the time of each census) to much more specialized information like the number of beds in Veterans Administration/Dept. of Veterans Affairs hospitals in the district or whether the district is coastal (see codebook for variable descriptions).  Most of the data are specific to each congressional district, but occasionally I have included state-wide data where district-level information was unavailable.

 

District data were collected once every ten years, with the exception of instances where district lines were redrawn mid-decade and new census data were available (these reapportioned districts are noted in each data file where they appear).  Normally, redrawn districts become effective at start of the new congressional term immediate following the availability of the new census data (usually in the third year of each decade [e.g., 1953, 1963, etc.,).  In several instances variables do not exist for every decade do to lack of availability of information.  I have included a table that lists which variables exist in each decade.

 


Data sources

A large portion of the data come from the decennial census.  Data from the 1960-1990 censuses were aggregated by congressional district in the Congressional Data Books and associated data files.  For the prior decades, county-level census data were aggregated into congressional districts (see description below).

 

Several variables in this data set were not contained in the census and had to be collected individually.  The non-census data come from a variety of sources that are listed in the codebook.  When sites such as military installations or VA hospitals needed to be placed into their appropriate districts, this was accomplished using an assortment of census Congressional District Atlases; Congressional Directories; Congressional Quarterly’s Congressional Districts in the 1970s, 1980s, and 1990s; Rand McNally Road Atlases; military road atlases; U.S. Geological Survey maps; and maps of congressional districts in the 1940s and 1950s provided from a private collection.

 


Aggregation Procedure for 1940s and 1950s Census Data

Data from the 1940 and 1950 censuses were available only at the county-level.  To construct district-level data for “whole-county” districts (districts entirely composed of one or more whole counties), county-level information was aggregated into congressional districts using a combination of maps and descriptions of congressional district boundaries published in the Congressional Directories and Kenneth Martis’s, The Historical Atlas of United States Congressional Districts, 1789-1983 (New York: Free Press, 1982).

 

For counties that were divided into several congressional districts (primarily those comprising large cities), or partitioned into districts that contained part of more than one county, I employed a system of data aggregation that utilized the Census’s Congressional District Atlas for 1960 (which contained the boundaries of districts for the pre-1962 districts) and Michael Dubin’s detailed maps of congressional districts.  Urban counties that contained multiple congressional districts were divided geographically and demographically according to the respective number of districts.  For example, the proportion unemployed in Los Angeles County in the 1950s was utilized as the proportion unemployed in the 15th through 26th congressional districts of California, which encompassed the county in that decade.  In several instances counties were divided among two congressional districts, but these county portions only made up part of an entire district.  In such cases the sum of the population of all the other counties in the district was subtracted from the population of the entire district (as provided in the Congressional Directory) to estimate the population contained in the divided county, and thus the proportion of the county within that district.

Preferred Citation
Adler, E. Scott.  “Congressional District Data File, [congressional term].” University of Colorado, Boulder, CO.


Please feel free to contact me if you have any questions, suggestions, or you identify errors or inconsistencies in the data: esadler@colorado.edu.

Table of Variables by Congressional Term


Data Download

Cong.   Year       File     

78th1943-44fin78.csv
79th
1945-46 
fin79.csv
80th1947-48
fin80.csv
81st   
1949-50
fin81.csv
82nd
1951-52
fin82.csv
83rd
1953-54
fin83.csv
84th   
1955-56
fin84.csv
85th
1957-58
fin85.csv
86th
1959-60
fin86.csv
87th
1961-62
fin87.csv
88th
1963-64
fin88.csv
89th
1965-66
fin89.csv
90th
1967-68
fin90.csv
91st
1969-70
fin91.csv
92nd
1971-72
fin92.csv
93rd   
1973-74
fin93.csv
94th
1975-76
fin94.csv
95th
1977-78
fin95.csv
96th
1979-80
fin96.csv
97th
1981-82
fin97.csv
98th
1983-84
fin98.csv
99th
1985-86
fin99.csv
100th
1987-88
fin100.cs
101st
1989-90
fin101.cs
102nd
1991-92fin102.cs
103rd  
1993-94
fin103.cs
104th1995-96
fin104.cs
105th
1997-98
fin105.cs